On-device ML

TensorFlow Lite 1.0 brings improvements to mobile machine learning

Sarah Schlothauer
© Shutterstock / denvitruk

Machine learning for mobile and Internet of Things devices just got easier. With the latest updates to TensorFlow Lite 1.0, ML heads towards your smart phone and smart home. See what new things the TensorFlow Dev Summit 2019 brings to the table.

It’s an understatement to say that TensorFlow reigns supreme when it comes to open source machine learning projects. Machine learning helps us create tools with a practically limitless range of uses.

When it comes to on-device machine learning, TensorFlow Lite is the next lightweight step. It takes machine learning down a mobile-friendly path. On March 6, 2019 at the TensorFlow Dev Summit in Sunnyvale, CA, developers were treated to an introduction for TensorFlow Lite 1.0.

Smart new features

TensorFlow Lite is an open source framework for deploying machine learning models on mobile and IoT devices.

SEE ALSO: Data nightmares: The tangible costs of bad data

Why is it necessary? The guide explains the reasoning behind this mobile-focused framework:

Machine Learning is changing the computing paradigm, and we see an emerging trend of new use cases on mobile and embedded devices. Consumer expectations are also trending toward natural, human-like interactions with their devices, driven by the camera and voice interaction models…We believe the next wave of machine learning applications will have significant processing on mobile and embedded devices.

TensorFlow guide

TensorFlow Lite allows for machine learning on devices as small as the Raspberry Pi, microcontrollers, or ARM64. (Currently, the release aimed at microcontrollers is in its experimental stage.)

According to an article from VentureBeat, TensorFlow Lite 1.0’s improvements include “selective registration and quantization during and after training for faster, smaller models. Quantization has led to 4 times compression of some models.”

This is all great news for on-device machine learning and points to even more improvements to come!

Oh, did we mention: TensorFlow Lite is even going to outer space. Out of this world!

Begin your ML journey

Ready? Here are some helpful TensorFlow Lite links to get you started:

For Android devices, installation takes only a few moments. You can simply download the pre-built binary APK and test out the demo.

For iOS users, there are a few prerequisites, but the tutorial is still fairly simple to get you up and running.

Keeping up with machine learning

Don’t forget to also stay updated with the rest of the happenings from the Dev Summit 2019!

SEE ALSO: Machine learning meets math: Solve differential equations with new Julia library

Plenty of machine learning announcements dropped, including news about the new and improved TensorFlow version 2.0 alpha release. Version 2.0 plans to make usage simpler and easier for use, referring to community offered criteria for improvements. Experts and beginners alike will find a variety of enhancements.

Find out how you can take the early preview for a test drive and see what the newest features bring.

View the livestreams on YouTube and transport yourself directly to sunny California while catching up on all the news.

Sarah Schlothauer

Sarah Schlothauer

All Posts by Sarah Schlothauer

Sarah Schlothauer is the editor for She received her Bachelor's degree from Monmouth University, West Long Branch, New Jersey. She currently lives in Frankfurt, Germany with her husband and cat where she enjoys reading, writing, and medieval reenactment. She is also the editor for Conditio Humana, an online magazine about ethics, AI, and technology.

Inline Feedbacks
View all comments